CN110008602A - A road network selection method considering multi-feature coordination under large scale - Google Patents

Abstract

The invention discloses a large-scale road network selection method considering multi-feature coordination, which includes constructing point, arc and polygon topologies for the road network, and recognizing meshes therein; meanwhile, generating stroke connections by considering road semantics, geometry and topological features, And identify the terminal arc and the terminal mesh; determine the mesh density threshold and the road stroke connection importance threshold; divide the road stroke type according to the terminal arc and the terminal mesh; and so on. The advantages are: by using this road network selection method, the road network selection can be completed by taking into account the road connectivity, integrity, and coordination of road network characteristics and local density multi-features in the process of road network selection; when selecting large-scale roads, It can better maintain the spatial characteristics of the road network; when selecting large-scale roads, it can better maintain the connectivity and integrity of the road network, and while taking into account the connectivity of the road network, it can well generalize the road. road network structure.

Description

A road network selection method considering multi-feature coordination under large scale

technical field

The invention relates to the technical field of geographic cartography, and in particular, to a road network selection method considering multi-feature coordination under a large scale.

Background technique

The road network on the map is an objective construction of the connection and distribution of the road network in the real geographic world, and is the skeleton element of the map. Usually, the road network has many levels, complex relationships, and a network. Therefore, the automatic synthesis of the road network has always been a difficult problem. In the process of road network selection, the selected focus depends on the scale span. However, the existing studies have not limited the comprehensive scale range for which the method is applicable. For the automatic map synthesis of urban large-scale (greater than 1:100,000) road networks, The construction of the road network is very delicate. Therefore, in the process of automatic comprehensive selection, it is necessary to consider the connectivity and integrity of the road itself, as well as the network characteristics and density characteristics of the overall road network.

The selection process of road network includes two aspects: how much to choose and which to choose. When the scale changes, the spatial distribution characteristics of the selection result completely depend on these two elements. Among them, the former is the problem of quota selection, which can generally be solved by the square root model; the latter is the problem of structured and optimal selection, which has always been a research hotspot. In the existing research, the selection method based on graph theory has laid a foundation for organizing road network data and taking into account the topology constraints of the road network. However, this method is difficult to realize the structured selection of the road network. The existing road network selection methods are as follows: First, by introducing the good continuation principle in Gestalt visual perception, the road sections are connected into strokes as the selection object, and the selection is completed according to the importance of the stroke to ensure the road network. Connectivity; second, calculate the importance of stroke, and use the length index to evaluate the importance of stroke, but the evaluation index is too single; third, consider the length of the stroke, the degree of connectivity and the average density of the included arcs, and add the stroke to the road. The second and third methods can effectively simulate the visual length of the road in manual selection, maintaining the connectivity of the road while considering the integrity of the road target, namely It can identify primary and secondary roads. However, the selection of secondary roads is relatively rough, resulting in the loss of road network network characteristics and local density characteristics of the road network. Fourth, the mesh density in road data reflects the local area. The density of roads in the area is obtained, and the density threshold is obtained to determine the selection rate. This method well maintains the characteristics of the road network in terms of density, topology, geometry and semantics. road section, disrupting road network connectivity.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a large-scale road network selection method that considers multi-feature coordination, so as to solve the aforementioned problems in the prior art.

In order to achieve the above object, the technical scheme adopted in the present invention is as follows:

A road network selection method considering multi-feature coordination under a large scale, comprising the following steps:

S1. Construct point, arc, and polygon topology for the road network, and identify the meshes therein; at the same time, consider road semantics, geometry and topological features to generate stroke connections, and identify distal arcs and distal meshes;

S2. Determine the mesh density threshold and the road stroke connection importance threshold;

S3. Divide the stroke type of the road according to the terminal arc and the terminal mesh;

S4. Determine whether each stroke connection in the divided road stroke type is a stroke connection containing a terminal mesh; if not, execute step S5; if so, execute step S7.

S5. Calculate the importance of the road stroke connection according to the road stroke type;

S6, determine whether the importance of the road stroke connection is less than the stroke connection importance threshold, if so, delete the road stroke connection, if not, keep the road stroke connection;

S7, collecting the stroke connections containing the terminal mesh to obtain the terminal mesh set;

S8. Classify the terminal mesh set according to the type of road stroke contained in the mesh, process the identified terminal mesh larger than the mesh density threshold, strip out the terminal mesh with the highest density and its associated road stroke set, and compare the importance of road strokes. Get the road stroke with the least importance;

S9, delete the road stroke with the least importance; and judge whether a hanging arc segment will be generated, if no hanging arc segment is generated, delete the road stroke, and merge the topological polygons on the left and right sides of the road stroke to generate a new mesh; If the arc segment is suspended, delete the end segment of the road stroke in the end mesh, and merge the topological polygons on the left and right sides of the segment to generate a new mesh;

S10. Determine whether the terminal mesh currently being processed is the last terminal mesh in the terminal mesh set that is greater than the mesh density threshold, and if so, output a new mesh, and if not, return to step S8.

Preferably, the identification process of the terminal arc segment in step S1 is to identify the arc segment in the road stroke connection with the number of intersections of all arc segments in the road stroke connection less than 2, and call the arc segment the terminal in the road stroke connection Arc segment; at the same time, it is recognized that there is a closed arc segment with the same start and end nodes, and the closed arc segment also belongs to the end arc segment.

Preferably, the identification process of the terminal mesh in step S1 is to identify the road mesh according to the topology relationship of the road network, and the mesh containing the terminal arc in the road stroke connection is called the terminal mesh.

Preferably, the determination process of the mesh density threshold in step S2 is to determine through the relationship between the mesh density of the same grade and the number of meshes.

Preferably, the mesh density is equal to the ratio of the total length of the road in the smallest area containing the mesh to the mesh area.

Preferably, the process of determining the importance threshold of the road stroke connection in step S2 is to determine through the visually distinguishable minimum distance on the map and the target scale.

Preferably, the method for dividing the road stroke type in step S3 includes the following contents:

S301, the set of other road strokes connected to the first point of the road stroke(S _i ) is marked as StartV(S _i ); the set of other road strokes connected to the end point of the road stroke(S _i ) is EndV(S _i ); The number of terminal arcs of road stroke(S _i ) is BurrN(S _i ); the number of road meshes associated with the terminal arcs of road strokeS _i is Net(L _i );

S302, by judging the above four parameters, the road stroke is divided into the following four categories:

Type I road stroke: Net(L _i )=0.

Type II road stroke: intersection[StartV(S _i ), EndV(S _i )]>0 and BurrN(S _i )=1 and Net(L _i )>0.

Class III road stroke: intersection[StartV(S _i ), EndV(S _i )]>0 and BurrN(S _i )>1 and Net(L _i )>0.

Class IV road stroke: intersection[StartV(S _i ), EndV(S _i )]=0 and Net(L _i )>0.

Preferably, the sorting order in step S8 is as follows: the meshes containing road-like strokes are first, the meshes containing road-like strokes second, and the meshes containing road-like strokes last.

The beneficial effects of the present invention are as follows: 1. The road network selection can be completed in consideration of road connectivity, integrity, and the coordination of road network network characteristics and local density multi-characteristics. 2. When selecting large-scale roads, the spatial characteristics of the road network can be better maintained. 3. When selecting large-scale roads, the connectivity and integrity of the road network can be better maintained, and the road network structure of the road can be well summarized while taking into account the connectivity of the road network.

Description of drawings

1 is a flowchart of a method for selecting a road network in an embodiment of the present invention;

2 is a schematic diagram of a distal arc segment and a distal mesh in an embodiment of the present invention;

3 is a comparison of estimated secondary road mesh density distributions of mesh density thresholds in an embodiment of the present invention;

Fig. 4 is the mesh density distribution comparison of the estimated main road mesh density threshold in the embodiment of the present invention;

5 is a schematic diagram of a road stroke classification in an embodiment of the present invention;

6 is a 1:50,000 standard drawing in an embodiment of the present invention;

7 is a stroke-based road network selection result in an embodiment of the present invention;

8 is a mesh-based road network selection result in an embodiment of the present invention;

FIG. 9 is a selection result of using the road network selection method of the present invention in an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

As shown in Figure 1, the present invention provides a large-scale road network selection method considering multi-feature coordination, including the following steps:

S1. Construct point, arc, and polygon topology for the road network, and identify the meshes therein; at the same time, consider road semantics, geometry and topological features to generate stroke connections, and identify distal arcs and distal meshes;

S2. Determine the mesh density threshold and the road stroke connection importance threshold;

S3. Divide the stroke type of the road according to the terminal arc and the terminal mesh;

S4. Determine whether each stroke connection in the divided road stroke type is a stroke connection containing a terminal mesh; if not, execute step S5; if so, execute step S7.

S5. Calculate the importance of the road stroke connection according to the road stroke type;

S6, determine whether the importance of the road stroke connection is less than the stroke connection importance threshold, if so, delete the road stroke connection, if not, keep the road stroke connection;

S7, collecting the stroke connections containing the terminal mesh to obtain the terminal mesh set;

S8. Classify the terminal mesh set according to the type of road stroke contained in the mesh, process the identified terminal mesh larger than the mesh density threshold, strip out the terminal mesh with the highest density and its associated road stroke set, and compare the importance of road strokes. Get the road stroke with the least importance;

S9, delete the road stroke with the least importance; and judge whether a hanging arc segment will be generated, if no hanging arc segment is generated, delete the road stroke, and merge the topological polygons on the left and right sides of the road stroke to generate a new mesh; If the arc segment is suspended, delete the end segment of the road stroke in the end mesh, and merge the topological polygons on the left and right sides of the segment to generate a new mesh;

S10. Determine whether the terminal mesh currently being processed is the last terminal mesh in the terminal mesh set that is greater than the mesh density threshold, and if so, output a new mesh, and if not, return to step S8.

In this embodiment, the hanging arc segment is an arc in which a part is connected and the remaining part is not connected.

In this embodiment, by adopting the above method, the road network selection can be completed in consideration of road connectivity, integrity, and coordination of road network network characteristics and local density multi-features during the road network selection process.

Example 1

As shown in FIG. 2 , this embodiment explains step S1. The road network selection method of the present invention needs to consider road semantics, geometry and topological features to generate stroke connections, and to identify peripheral features such as peripheral arcs and peripheral meshes.

In this embodiment, stroke is derived from the principle of good continuity in the Gestalt cognitive principle, and the concept is generated from the idea of drawing a curve segment with one stroke. Construct road network point, line, surface topology, and form road stroke connections according to information such as arc semantics, direction, length, etc., as shown in Figure ₂ , road stroke connections S1, _S2 , _S3 , _S4 , _S5 , _S6 . For the terminal arc, if the number of intersections between an arc in the road stroke connection and all arcs in the road stroke connection is less than 2, the arc is called the terminal arc in the road stroke connection. Including arcs AB and DE in S1, arcs FG and _IJ in S2, _arcs KL and NO in S3, _arcs BG and LP in S4, and _{arcs CH} and MQ in S5 , the arc segments DI and IN in _S6 . At the same time, for a closed arc segment with the same start and end nodes, the closed arc segment also belongs to the terminal arc segment.

In this embodiment, for the terminal meshes, the road meshes are identified according to the topology relationship of the road network, such as meshes I, II, III, and IV, and the meshes containing the terminal arcs in the stroke connection of the road are called terminal meshes, such as meshes I, III, and IV. II and IV.

Embodiment 2

As shown in FIG. 3 and FIG. 4 , this embodiment explains step S2. In the road network selection method of the present invention, it is necessary to calculate and determine two parameters, a mesh density threshold (TN) and a road stroke connection importance threshold (TS). Follow-up to carry out the selection of the road stroke.

In this embodiment, whether to select the mesh in the road needs to be determined by comparing the mesh density of the road with the mesh density threshold (TN); :

D=P/A

Among them, D is the mesh density, P is the total length of the road section on the mesh boundary, and A is the area of the mesh.

In this embodiment, the mesh density threshold (TN) can usually be determined by a method based on statistical analysis, and the density threshold is determined by analyzing the relationship between the mesh density of the same level and the number of meshes before and after the sample image is synthesized. The source scale is 1:10,000 and the target scale is 1:50,000 as an example to illustrate that the roads are divided into two types: main roads and secondary roads, and the meshes are divided into meshes composed of main roads and meshes composed of secondary roads. The curves in Fig. 3 and Fig. 4 represent the relationship between the density value and the number of meshes whose density is this value, respectively representing the comparison of the density distribution of the two types of meshes at different scales. It can be seen from Figure 3 that the density value of 0.012m/m is the dividing line between the two intervals, and the mesh with a density greater than 0.012 needs to be selected at a scale of 1:50,000; the distribution curve in Figure 4 shows that the mesh density distributions of the two main roads are almost identical , indicating that the main road is almost not abandoned at the scale of 1:50,000, then 0.012 can be selected as the mesh density threshold (TN) at the scale of 1:50,000.

In this embodiment, whether the stroke in the road is selected needs to be determined by comparing the importance of the road stroke with the road stroke connection importance threshold (TS). The road stroke calculation method is different. For road strokes with terminal meshes, the stroke importance is calculated according to the following formula:

I=BC×L

where I is stroke importance; BC is stroke betweenness centrality; L is stroke length.

For road strokes that do not contain terminal meshes, the stroke importance is calculated according to the following formula.

I=(1+N)×L

where I is the stroke importance; N is the stroke connectivity; L is the stroke length.

The road stroke connection importance threshold (TS) was determined by the visually distinguishable minimum distance on the map and the target scale. Usually, cartography experts believe that the visually distinguishable distance on the map is 0.4mm, then under the target scale (1: Scale _target ), the road stroke connection importance threshold (TS) is calculated according to the following formula:

Ts=0.4×Scale _target

Embodiment 3

As shown in FIG. 5 , in this embodiment, the road network selection method in the present invention needs to divide the road stroke type according to the set of road strokes associated with the head and end points of the stroke, the number of arc segments at the end and the number of meshes at the end. The set of other road strokes connected to the first point of the road stroke(S _i ) is recorded as StartV(S _i ); the set of other road strokes connected to the end point of the road stroke(S _i ) is called EndV(S _i ); the road stroke The number of tip arcs of (S _i ) is BurrN(S _i ); the number of road meshes associated with the tip arcs of the road stroke(S _i ) is Net(L _i );

S302, by judging the above four parameters, the road stroke is divided into the following four categories:

Type I road stroke: Net(L _i )=0.

Type II road stroke: intersection[StartV(S _i ), EndV(S _i )]>0 and BurrN(S _i )=1 and Net(L _i )>0.

Class III road stroke: intersection[StartV(S _i ), EndV(S _i )]>0 and BurrN(S _i )>1 and Net(L _i )>0.

Class IV road stroke: intersection[StartV(S _i ), EndV(S _i )]=0 and Net(L _i )>0. As shown in Fig. 5 , the strokes of type I roads include S ₁ , S ₂ , S ₃ , S ₄ , S ₉ , S ₁₁ , S ₁₂ , S ₁₃ , S ₁₄ , and S ₁₅ , and the strokes of type II roads include S ₈ , III Class road strokes have S ₅ , and class IV road strokes have S ₆ , S ₇ , and S ₁₀ .

Embodiment 4

As shown in Figures 6 to 9, three methods are used for road selection in this embodiment, wherein Figure 6 is a 1:50,000 standard map, Figures 7, 8 and 9 respectively use stroke-based road network selection The method, the mesh-based road network selection method and the method of the present invention, the road selection result is from 1:10,000 synthesis to 1:50,000. Compared with the results of the standard map in Figure 6, for the road in the rectangle A, the results of the stroke-based road network selection method, in Figure 7, the road segment a at the end is retained, but the connected road segment b is lost, resulting in road connectivity. Destroyed; the results of the mesh-based road network selection method, the road section b is retained in FIG. 8, but the road section a is lost, causing the road integrity to be destroyed; the selection method of the present invention, the road sections a and b are retained in FIG. In this way, the connectivity and integrity of the road network are better maintained. In addition, for the road in rectangle B, affected by mesh aggregation, the stroke-based road network selection method cannot detect the complex structure in Fig. 7, resulting in the loss of the original structure and the appearance of hanging arcs; the mesh-based road network selection method Although the connectivity of the road network is considered in Fig. 8, the structure has changed significantly; the selection method of the present invention, the main road here is well extracted in Fig. 9, while taking into account the connectivity of the road network , which gives a good overview of the road network structure there.

By adopting the above-mentioned technical scheme disclosed by the present invention, the following beneficial effects are obtained:

The invention provides a road network selection method considering multi-feature coordination under a large scale. By adopting this method to select a road network from an inverted database, the road connectivity, integrity, and coordination of road network network characteristics and local density multi-characteristics can be considered. Complete road network selection; when selecting large-scale roads, it can better maintain the spatial characteristics of the road network; when selecting large-scale roads, it can better maintain the connectivity and integrity of the road network, and take into account the road network. At the same time of network connectivity, the road network structure of the road is well summarized.

The above are only the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, several improvements and modifications can be made. It should be regarded as the protection scope of the present invention.

Claims (8)

1. a road network selection method considering multi-feature coordination under a large scale is characterized in that: comprise the steps, S1. Construct point, arc, and polygon topology for the road network, and identify the meshes therein; at the same time, consider road semantics, geometry and topological features to generate stroke connections, and identify distal arcs and distal meshes; S2. Determine the mesh density threshold and the road stroke connection importance threshold; S3. Divide the stroke type of the road according to the terminal arc and the terminal mesh; S4. Determine whether each stroke connection in the divided road stroke type is a stroke connection containing a terminal mesh; if not, execute step S5; if so, execute step S7. S5. Calculate the importance of the road stroke connection according to the road stroke type; S6, determine whether the importance of the road stroke connection is less than the stroke connection importance threshold, if so, delete the road stroke connection, if not, keep the road stroke connection; S7, collecting the stroke connections containing the terminal mesh to obtain the terminal mesh set; S8. Classify the terminal mesh set according to the type of road stroke contained in the mesh, process the identified terminal mesh larger than the mesh density threshold, strip out the terminal mesh with the highest density and its associated road stroke set, and compare the importance of road strokes. Get the road stroke with the least importance; S9, delete the road stroke with the least importance; and determine whether a hanging arc segment will be generated, if no hanging arc segment is generated, delete the road stroke, and merge the topological polygons on the left and right sides of the road stroke to generate a new mesh, and execute S10; if a hanging arc segment is generated, delete the terminal road segment of the road stroke in the terminal mesh, merge the topological polygons on the left and right sides of the road segment, generate a new mesh, and execute S10; S10. Determine whether the terminal mesh currently being processed is the last terminal mesh in the terminal mesh set that is greater than the mesh density threshold, and if so, output a new mesh, and if not, return to step S8. 2. the road network selection method considering multi-feature coordination under the large scale according to claim 1, is characterized in that: the identification process of tip arc segment in step S1 is, identify all arcs in road stroke connection with this road stroke connection The arc segment whose number of intersections is less than 2 is called the end arc segment in the stroke connection of the road; at the same time, it is identified that there is a closed arc segment with the same start and end nodes, and the closed arc segment also belongs to the end arc segment. 3. the road network selection method considering multi-feature coordination under the large scale according to claim 1, is characterized in that: the identification process of tip mesh in step S1 is, according to road network topology relationship, identify road mesh, will contain road stroke The meshes that connect the middle and distal arc segments are called distal meshes. 4. the road network selection method considering multi-feature coordination under the large scale according to claim 1 is characterized in that: the determination process of the mesh density threshold value in the step S2 is to determine by the relationship between the mesh density of the same grade and the number of meshes . 5 . The road network selection method considering multi-feature coordination under a large scale according to claim 4 , wherein the mesh density is equal to the ratio of the total length of the road in the smallest area containing the mesh to the mesh area. 6 . 6. the road network selection method considering multi-feature coordination under the large scale according to claim 1, is characterized in that: in step S2, the determination process of road stroke connection importance threshold value is, through the minimum distance visually distinguishable on the graph and target scale. 7. the road network selection method in consideration of multi-feature coordination under the large scale according to claim 1, is characterized in that: the division method of road stroke type in step S3 comprises the following content, S301, the set of other road strokes connected to the first point of the road stroke(S _i ) is marked as StartV(S _i ); the set of other road strokes connected to the end point of the road stroke(S _i ) is EndV(S _i ); The number of terminal arcs of road stroke(S _i ) is BurrN(S _i ); the number of road meshes associated with the terminal arcs of road strokeS _i is Net(L _i ); S302, by judging the above four parameters, the road stroke is divided into the following four categories: Type I road stroke: Net(L _i )=0. Type II road stroke: intersection[StartV(S _i ), EndV(S _i )]>0 and BurrN(S _i )=1 and Net(L _i )>0. Class III road stroke: intersection[StartV(S _i ), EndV(S _i )]>0 and BurrN(S _i )>1 and Net(L _i )>0. Class IV road stroke: intersection[StartV(S _i ), EndV(S _i )]=0 and Net(L _i )>0. 8. the road network selection method considering multi-feature coordination under the large scale according to claim 1, is characterized in that: the classification sequence in step S8 is, firstly is the mesh that contains class road stroke, and secondly is the mesh that contains class road stroke. The mesh, and finally the mesh that contains the road-like stroke.